I’ve noticed a trend lately of large articles. Large articles that I end up not reading because I don’t know where the article is going. If you’re writing a full length article, starting with an abstract would be helpful, IMO.
Thanks for the feedback; I didn’t write these for LW (posting them because it seems like it can do only good; there doesn’t seem to be enough interest here to be worth putting much effort into translation). So I don’t expect them to fit in particularly well; but small changes to make posts more accessible are still good, particularly when it takes the form of good general advice for writing.
Right now I think it should be clear after the first 2-3 paragraphs where the piece is going, and I tried to make it very explicit after the first 5 paragraphs. Is the issue:
5 paragraphs is much too long, and the intro/title is not enough to motivate you to get through the first 5 paragraphs.
It’s not clear from the first 5 paragraphs where the post is going.
An abstract should be explicitly presented as such.
There should be a concise statement of the form “where is this post going? Here” rather than being spread out over dozens of words.
The answer the post arrives at should be given in the abstract, in addition to the question (I don’t know how I would do this one in this case).
(I could ask the same questions about the first 2-3 paragraphs; is 2-3 too many, is it not yet clear, should it be more explicitly called out?)
First, I’d note that in response to a comment that I found it unclear where the article is going, and suggesting that you clarify it with an abstract, you responded asking for clarifications about my comment, and not a word clarifying the point of your article.
I have no doctrine on how many paragraphs one must use to convey what one is doing, or to hold a reader’s interest. My interest was not held, and it wasn’t clear to me where you were going.
A long post requires more time and more reading. After a few paragraphs, if the return on reading seems low, I’ll stop. I felt the return was low. I stopped.
The answer the post arrives at should be given in the abstract, in addition to the question (I don’t know how I would do this one in this case).
As for abstracts, I identify the problem. Identify the conclusions, if any. More generally, identify the payoff to the reader of reading the article. Motivate the reader.
This is a general issue with long posts. A longer posts requires greater investment, and so greater motivation. Motivate the reader.
The post is an answer to the question “what framework is appropriate for practical inductive reasoning?”
I was asking: is this unclear from the first few paragraphs, is the first few paragraphs too long to take (it sounds like you read the first few paragraphs, so this isn’t it), is this an insufficiently precise description of what the post is about, is this an unclear description of what the post is about? Is more necessary for motivation? I am happy to only appeal to people who have thought about the issue before.
I identify the problem in the first paragraph. I can’t identify the conclusion concisely. The payoff to the reader is a proposed approach to the problem, which should be useful to them if they find the problem problematic and are interested in new perspectives. It should not be useful to others.
Going by the first 3 paragraphs, it was unclear to me whether you would be writing about climate change or a particular related problem in inference, of estimating the probability of low likelihood, high cost events as extrapolated from historical data.
It did not seem in any way so general as
The post is an answer to the question “what framework is appropriate for practical inductive reasoning?”
And looking through the rest of the post, I don’t think the post as a whole is that general either. More like decision theory for low likelihood, high cost events as extrapolated from historical data.
I’ve noticed a trend lately of large articles. Large articles that I end up not reading because I don’t know where the article is going. If you’re writing a full length article, starting with an abstract would be helpful, IMO.
Thanks for the feedback; I didn’t write these for LW (posting them because it seems like it can do only good; there doesn’t seem to be enough interest here to be worth putting much effort into translation). So I don’t expect them to fit in particularly well; but small changes to make posts more accessible are still good, particularly when it takes the form of good general advice for writing.
Right now I think it should be clear after the first 2-3 paragraphs where the piece is going, and I tried to make it very explicit after the first 5 paragraphs. Is the issue:
5 paragraphs is much too long, and the intro/title is not enough to motivate you to get through the first 5 paragraphs.
It’s not clear from the first 5 paragraphs where the post is going.
An abstract should be explicitly presented as such.
There should be a concise statement of the form “where is this post going? Here” rather than being spread out over dozens of words.
The answer the post arrives at should be given in the abstract, in addition to the question (I don’t know how I would do this one in this case).
(I could ask the same questions about the first 2-3 paragraphs; is 2-3 too many, is it not yet clear, should it be more explicitly called out?)
First, I’d note that in response to a comment that I found it unclear where the article is going, and suggesting that you clarify it with an abstract, you responded asking for clarifications about my comment, and not a word clarifying the point of your article.
I have no doctrine on how many paragraphs one must use to convey what one is doing, or to hold a reader’s interest. My interest was not held, and it wasn’t clear to me where you were going.
A long post requires more time and more reading. After a few paragraphs, if the return on reading seems low, I’ll stop. I felt the return was low. I stopped.
As for abstracts, I identify the problem. Identify the conclusions, if any. More generally, identify the payoff to the reader of reading the article. Motivate the reader.
This is a general issue with long posts. A longer posts requires greater investment, and so greater motivation. Motivate the reader.
The post is an answer to the question “what framework is appropriate for practical inductive reasoning?”
I was asking: is this unclear from the first few paragraphs, is the first few paragraphs too long to take (it sounds like you read the first few paragraphs, so this isn’t it), is this an insufficiently precise description of what the post is about, is this an unclear description of what the post is about? Is more necessary for motivation? I am happy to only appeal to people who have thought about the issue before.
I identify the problem in the first paragraph. I can’t identify the conclusion concisely. The payoff to the reader is a proposed approach to the problem, which should be useful to them if they find the problem problematic and are interested in new perspectives. It should not be useful to others.
Going by the first 3 paragraphs, it was unclear to me whether you would be writing about climate change or a particular related problem in inference, of estimating the probability of low likelihood, high cost events as extrapolated from historical data.
It did not seem in any way so general as
And looking through the rest of the post, I don’t think the post as a whole is that general either. More like decision theory for low likelihood, high cost events as extrapolated from historical data.